Date: Sat, 15 Feb 2020 21:12:27 +0300 From: Lev Serebryakov <lev@FreeBSD.org> To: David Wolfskill <david@catwhisker.org> Cc: current@freebsd.org Subject: Re: CURRENT panciks right after kernel load (r357966) Message-ID: <1256487134.20200215211227@serebryakov.spb.ru> In-Reply-To: <20200215180253.GW1238@albert.catwhisker.org> References: <241831336.20200215205415@serebryakov.spb.ru> <20200215180253.GW1238@albert.catwhisker.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Hello David, Saturday, February 15, 2020, 9:02:53 PM, you wrote: >> CURRENT r357966 on amd64, on host with 4G of memory (and Atom CPU) panics >> with page fault right after kernel load (before any kernel output). >> >> Stacktrace (typed from photo of screen, all trap-related functions are >> omitted): >> >> uma_zalloc_arg() >> malloc() >> sysctl_handle_string() >> sysctl_netisr_dispatch_policy() >> sysctl_root_handler_locked() >> sysctl_register_oid() >> sysctl_register_all() >> mi_startup() >> ..... > My last CURRENT build was at r357959 (and which booted without issue -- > though the systems in question don't use ZFS); only commits I see to > head after r357959 are r357962 & r357967. I've tried r357763 and it panics exactly same way. My system doesn't have ZFS too (it is NanoBSD image for small router). I didn't update this "firmware" for almost a year, and don't have results for revision between r349299 (works) and r357763 (panics). It is too much for bi-sect :-( I'll try r357508, which works for me on other host, but I suspect it is HW-related and not universal. But, again, r349299 works rock-solid. -- Best regards, Lev mailto:lev@FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1256487134.20200215211227>